What this is about
A change is planned in weeks, a habit grows over months. Those two clocks do not run together, and that explains why many changes sink away after everything seemed to be going well. This insight is about the slow clock: how behaviour becomes automatic, how long that takes, and how you see when it has got there.
How behaviour becomes a habit
A habit is behaviour that no longer asks for a decision. You do not think about how you log into the system you have used for years, your hands do it. New behaviour only gets there one way: through repetition, always in the same situation. Every time someone does the new thing at the moment it belongs, the path gets a little deeper. Until the situation itself calls up the behaviour, with no choice involved any more.
Until that day there is a choice every single time. And a choice costs energy, certainly when the old path runs deeper than the new one. That is the vulnerable period: the new way can already work, and still lose to the old one every day, simply because the old one happens by itself.
A good two months, and often longer
Research into habit formation gives a rough figure: a good two months of consistent repetition on average before new behaviour becomes automatic. On average, with large differences: simple behaviour settles sooner, complex behaviour can take twice as long or more. For a way of working that involves several steps and has to happen under time pressure, you are better off counting in months than in weeks.
Lay that alongside the calendar of an average project. The intensive support around a roll-out lasts a few weeks. After that it winds down, because the roll-out succeeded. The arithmetic is sobering: the attention stops somewhere halfway through the period the behaviour needed. Not because anyone made a mistake, but because nobody had the second clock in the plan.
Skipping once breaks nothing
There is reassurance in the research as well: one missed time does not undo the build-up. Someone who falls back on the old way once in a busy week is not back to square one. The path already dug stays there. That is important to tell teams, because the misconception is paralysing: anyone who thinks one miss breaks everything gives up after that one miss.
What does break the build-up is a pattern. One exception is a ripple, an exception that comes back every week is a second path getting deeper. You guard that difference not with strictness but with attention: the exception may exist, as long as someone sees when it starts becoming a habit.
How you see that it has taken hold
Not by the calendar, but by the behaviour itself. Three signs are reliable. The first: the busiest day. Under pressure people reach for what comes naturally, so if the team does the new thing in the peak week as well, the path is deep enough. The second: the new colleague. Someone who has just started learns from what colleagues do, not from what the manual says. If the newcomer is taught the new way as "this is how we do it here", the way of working has become culture. The third: the language. As long as people talk about "the new system", it is new. The day it is simply called "the system", the change has become invisible, and that is exactly the point.
The calendar follows the behaviour, not the other way round
All of this leads to one practical rule: the end of the follow-up is not a date, but an observation. You stop paying extra attention when the signs say the behaviour has taken hold, not when the budget runs out. Sometimes that is earlier than planned, often later. Agree it that way up front and you avoid the classic situation where the follow-up ends because the project ends, in the middle of the vulnerable period.
How we go about it
We put the second clock in the plan. Alongside the project calendar sits a behaviour calendar: how many months of repetition this behaviour realistically needs, who watches it in that period, and which signs tell us it is over. The people watching have not been hired for it: they are the colleagues and managers who helped shape the change and are there anyway.
And we tell teams explicitly that a missed time is not a break. That one insight prevents more giving up than any amount of checking. Read more about how we work →
When this insight helps
When planning every change that rests on behaviour, and that is nearly all of them. And when evaluating a change in progress: if the support ends while the behaviour is still a daily choice, you now know what is at stake in the months after, and how little it costs to have someone watch it.